WordPress AI Failure Patterns: A Research and Classification Protocol

A WordPress AI failure catalog should preserve raw evidence and distinguish task-design, evidence, connection, permission, tool, model, implementation and verification failures instead of blaming every problem on the model.

AI is most useful here as an evidence organizer, comparison engine and drafting assistant. It can make a complex WordPress task easier to inspect, but it cannot create missing authority, certify facts it did not observe or silently convert a recommendation into permission to act.

In one sentence: A WordPress AI failure catalog should preserve raw evidence and distinguish task-design, evidence, connection, permission, tool, model, implementation and verification failures instead of blaming every problem on the model.

What this guide helps you accomplish

Build a reproducible failure taxonomy and incident corpus that supports product improvement, safer instructions and more accurate public guidance.

• A multi-layer failure taxonomy with decision rules.
• A sanitized incident record format linked to exact versions and tasks.
• Frequency, severity, detectability and recovery fields.
• A process for converting verified patterns into tests, documentation or product controls.

The finished artifact should be understandable by the person responsible for the decision and reproducible by someone who did not participate in the original prompt. A fluent answer is not enough. Every material conclusion needs a source, a scope and a verification path. When the evidence cannot establish something, the correct output is an explicit unknown or a testable hypothesis.

Evidence and inputs to prepare

• Failed benchmark runs, support cases and lab incidents.
• Raw sanitized prompts, tool calls, errors, state diffs and verification results.
• Exact WordPress, plugin, client, model and transport versions.
• Expected task, permission and evidence contracts.
• Reviewer dispositions and remediation evidence.

Before supplying evidence to an assistant, remove credentials, secret values and unrelated personal information. Preserve the identifiers, versions, timestamps, locale, units and source labels needed to interpret what remains. A screenshot without a URL, state or date may be useful context, but it is rarely sufficient authority for a production decision.

Do not begin with a broad request such as “review this,” “fix this” or “make it better.” Define the decision the work must support, the population included, the source that is authoritative for each field, the allowed operations and the actions that remain forbidden. The planning or research stage should use a local repository, isolated fixture or exported evidence and does not require production WordPress access.

Failure location is not failure cause

An assistant may produce the visible error because a task lacked evidence, a route was absent, a permission was correct or a fixture was invalid.

Unsafe success is a failure

A task that completes by exceeding scope, publishing without approval or inventing evidence should be classified as a failure even when the requested page exists.

Taxonomy must support action

Categories should lead to a better prompt, product control, test, permission rule, connection fix or documentation change.

Keep observation, inference and authority separate

A controlled review should distinguish at least four states:

1. Observed: directly present in a named record, file, response, rendered page or executed test.
2. Inferred: a plausible interpretation supported by evidence but not directly established.
3. Recommended: a proposed human decision or next action.
4. Authorized and verified: a separately approved change that was executed and then checked against acceptance criteria.

AI output usually begins in the first three states. It does not become authorized merely because it is detailed, internally consistent or technically convincing. Preserve this distinction in tables, reports, tickets and public case studies.

A safe workflow

1. Define the layers and decision rules before reviewing incidents.
2. Collect raw sanitized evidence with exact version and task context.
3. Separate observed event, user impact, detection and causal hypotheses.
4. Have independent reviewers classify a sample and resolve disagreements.
5. Measure recurrence, severity, detectability and recovery burden where data permits.
6. Link verified patterns to tests, documentation, product controls or open research.
7. Re-run relevant cases after changes.
8. Publish only aggregated, non-sensitive findings with explicit denominators and limits.

This sequence deliberately places accountable review between analysis and implementation. If a later stage needs broader access, create a new task, a new identity or an explicit permission change. Do not quietly upgrade the analytical identity because it reached a correct boundary.

Prompt recipe

Replace every value in square brackets before using the prompt. Do not paste passwords, API keys, authentication cookies, private customer records or unrelated personal information.

You are reviewing [TASK SCOPE] for [SITE, REPOSITORY OR DATASET] using only the supplied evidence.

Objective:
Build a reproducible failure taxonomy and incident corpus that supports product improvement, safer instructions and more accurate public guidance.

Return the following fields:
- Incident ID
- Task ID
- Observed event
- Expected outcome
- WordPress state
- Version set
- Failure layer
- Severity
- Detection
- Recovery
- Evidence
- Causal confidence
- Disposition

Rules:
1. Preserve raw evidence before classification.
2. Do not infer cause from the visible error alone.
3. Classify unsafe success as a failure.
4. Record reviewer disagreement and unknown causes.
5. Do not publish sensitive incident details.

For every finding:
- identify the exact source, record, URL, file, line, object ID, state or dataset row;
- preserve dates, versions, units, locale, identifiers and denominators;
- separate observation, inference, recommendation and unknown;
- state what evidence was not available;
- do not change WordPress, source code, commerce data, analytics, external systems or published content.

Why this prompt is structured this way

The prompt creates an evidence contract before asking for recommendations. It makes missing data visible, reduces the chance that a model will complete an incomplete record with plausible prose and produces an output that can be reviewed systematically. Structured fields also make it easier to compare repeated runs or hand an approved subset to a later implementation workflow.

A production implementation may add JSON schema, typed tool inputs or automated validation. Those mechanisms improve consistency, but they do not establish that the source evidence is true, complete or current. Human review and system-specific verification remain required.

Recommended access boundary

Use No WordPress access during the planning or research stage for the stage described in this guide. The exact capabilities available to an identity must come from the installed product version, the published coverage contract and the connection method actually in use.

What must remain outside this task

• Fabricated incident counts
• Security disclosure without review
• User blame
• Single-cause simplification
• Deleting unsuccessful benchmark runs

A refused action can be useful evidence that the control boundary is working. Do not respond to an expected refusal by granting a broad administrator account or Full Power. First determine whether the action belongs in the current mandate at all. If it does, create a separately authorized stage with the narrowest required capability.

How WP Agent Control fits

WP Agent Control can provide a dedicated WordPress identity and a bounded permission profile for the stages its installed version actually supports.

WP Agent Control is the controlled WordPress identity and permission layer. It is not the AI model, not a universal MCP server and not proof that every assistant, client or transport can reach every WordPress surface. The assistant, client, transport, WordPress identity, task permission and human approval are separate layers.

Full Power is a distinct administrative exception. It must never be presented as the ordinary continuation of Read Only, Draft, Content Editor or Publisher, and it must not be used merely to make an example, benchmark or workflow succeed after a correct refusal.

Verification checklist

• ☐ The task, population, period, environment and decision are explicit.
• ☐ Every material observation is linked to exact evidence or labelled as a hypothesis.
• ☐ Stable IDs, URLs, versions, dates, units, locales and denominators are preserved.
• ☐ Missing evidence and coverage limits remain visible.
• ☐ The analytical or research identity performed no prohibited mutation.
• ☐ A qualified owner reviewed security, accessibility, legal, commerce or release implications where applicable.
• ☐ Any implementation has a separate mandate, access level, backup and verification plan.
• ☐ Temporary identities, fixtures and sensitive evidence are revoked, reset or disposed of after the task.

Common failure modes

• Model monocause: Every incident is attributed to hallucination even when the task or permission contract was defective.
• Only visible failures counted: Unauthorized or unverifiable successes disappear from the catalog.
• Denominator loss: A frequent-sounding pattern is published without the number and type of runs observed.
• Post-fix closure without rerun: A documentation or code change is assumed to resolve the pattern without reproduction.

A recurring cross-cutting failure is permission drift: the initial task encounters a limit, and the operator broadens access before determining whether the missing operation is necessary, supported or safe. This destroys the evidence value of the refusal and makes later results difficult to attribute.

Research status and publication gate

This page defines a protocol, not a completed study. It contains no benchmark values, provider rankings, success rates or empirical conclusions.

Before public release, the study needs a pre-registered protocol, a frozen fixture, an approved budget, repeated runs, deterministic verification, reviewer rules and a sanitized evidence package. Any result must state its numerator, denominator, missing runs, exact version set and uncertainty. A later model, client, WordPress release or permission profile is a different treatment and should not inherit the earlier conclusion automatically.

Advanced note

A useful failure ledger connects mandate, evidence, execution, restitution and verification. This makes it possible to see whether a defect originated before the model was called, during tool execution or in the interpretation of the result.

Related guides

• How to Analyze WordPress Debug Logs with AI
• WordPress AI Refusal Study: Measuring Whether Access Controls Fail Safely
• How to Build a WordPress AI Task Coverage Matrix
• How to Document a Controlled WordPress AI Workflow Case Study

Next step

Continue with the most relevant supporting guide and use the access-level guide before any authenticated task. When temporary WordPress access is no longer needed, finish by revoking the identity.